feat: add Cloudflare Radar top 1m domains dataset

This commit is contained in:
MDLeom 2022-11-25 07:19:20 +00:00
parent c376e2a08f
commit 4bf534bdbc
No known key found for this signature in database
GPG Key ID: 32D3E28E96A695E8
3 changed files with 62 additions and 26 deletions

View File

@ -12,7 +12,7 @@ build_job:
stage: build
before_script:
- apk update && apk add brotli curl grep xmlstarlet
- apk update && apk add brotli curl grep jq xmlstarlet
script:
- sh src/script.sh
@ -46,7 +46,7 @@ cloudflare:
stage: deploy
before_script:
- apk update && apk add curl
- apk update && apk add curl
script:
- curl -X POST "https://api.cloudflare.com/client/v4/pages/webhooks/deploy_hooks/$CLOUDFLARE_BUILD_HOOK"

View File

@ -1,21 +1,21 @@
# Phishing URL Blocklist
- Formats
* [URL-based](#url-based)
* [Domain-based](#domain-based)
* [Hosts-based](#hosts-based)
* [Domain-based (AdGuard Home)](#domain-based-adguard-home)
* [URL-based (AdGuard)](#url-based-adguard)
* [URL-based (Vivaldi)](#url-based-vivaldi)
* [Dnsmasq](#dnsmasq)
* [BIND zone](#bind)
* [RPZ](#response-policy-zone)
* [Unbound](#unbound)
* [dnscrypt-proxy](#dnscrypt-proxy)
* [Tracking Protection List (IE)](#tracking-protection-list-ie)
* [Snort2](#snort2)
* [Snort3](#snort3)
* [Suricata](#suricata)
- [URL-based](#url-based)
- [Domain-based](#domain-based)
- [Hosts-based](#hosts-based)
- [Domain-based (AdGuard Home)](#domain-based-adguard-home)
- [URL-based (AdGuard)](#url-based-adguard)
- [URL-based (Vivaldi)](#url-based-vivaldi)
- [Dnsmasq](#dnsmasq)
- [BIND zone](#bind)
- [RPZ](#response-policy-zone)
- [Unbound](#unbound)
- [dnscrypt-proxy](#dnscrypt-proxy)
- [Tracking Protection List (IE)](#tracking-protection-list-ie)
- [Snort2](#snort2)
- [Snort3](#snort3)
- [Suricata](#suricata)
- [Compressed version](#compressed-version)
- [Reporting issues](#issues)
- [See also](#see-also)
@ -310,7 +310,7 @@ chmod 755 /etc/cron.daily/phishing-filter
Configure dnscrypt-proxy to use the blocklist:
``` diff
```diff
[blocked_names]
+ blocked_names_file = '/etc/dnscrypt-proxy/phishing-filter-dnscrypt-blocked-names.txt'
@ -407,7 +407,7 @@ chmod 755 /etc/cron.daily/phishing-filter
Configure Snort to use the ruleset:
``` diff
```diff
# /etc/snort/snort.lua
ips =
{
@ -448,7 +448,7 @@ chmod 755 /etc/cron.daily/phishing-filter
Configure Suricata to use the ruleset:
``` diff
```diff
# /etc/suricata/suricata.yaml
rule-files:
- local.rules
@ -475,13 +475,13 @@ All filters are also available as gzip- and brotli-compressed.
- Gzip: https://malware-filter.gitlab.io/malware-filter/phishing-filter.txt.gz
- Brotli: https://malware-filter.gitlab.io/malware-filter/phishing-filter.txt.br
*Snort 2 rule is only available in compressed format in pages.dev due to the platform's 25MB file size limit*
_Snort 2 rule is only available in compressed format in pages.dev due to the platform's 25MB file size limit_
## Issues
This blocklist operates by blocking the **whole** website, instead of specific webpages; exceptions are made on popular websites (e.g. `https://docs.google.com/`), in which webpages are specified instead (e.g. `https://docs.google.com/phishing-page`). Phishing webpages are only listed in [URL-based](#url-based) filter, popular websites are excluded from other filters.
*Popular* websites are as listed in the [Umbrella Popularity List](https://s3-us-west-1.amazonaws.com/umbrella-static/index.html) (top 1M domains + subdomains), [Tranco List](https://tranco-list.eu/) (top 1M domains) and this [custom list](src/exclude.txt).
_Popular_ websites are as listed in the [Umbrella Popularity List](https://s3-us-west-1.amazonaws.com/umbrella-static/index.html) (top 1M domains + subdomains), [Tranco List](https://tranco-list.eu/) (top 1M domains), [Cloudflare Radar](https://developers.cloudflare.com/radar/investigate/domain-ranking-datasets/) (top 1M domains) and this [custom list](src/exclude.txt).
If you wish to exclude certain website(s) that you believe is sufficiently well-known, please create an [issue](https://gitlab.com/malware-filter/phishing-filter/issues) or [merge request](https://gitlab.com/malware-filter/phishing-filter/merge_requests).
@ -504,6 +504,7 @@ Optional variables:
- `PHISHTANK_API`: Recommended if you intend to run [script.sh](src/script.sh) >5 times daily. Register an account at [phishtank.org](https://phishtank.org/developer_info.php) to generate an application key.
- `CLOUDFLARE_BUILD_HOOK`: Deploy to Cloudflare Pages.
- `NETLIFY_SITE_ID`: Deploy to Netlify.
- `CF_API`: Include Cloudflare Radar [domains ranking](https://developers.cloudflare.com/radar/investigate/domain-ranking-datasets/). [Guide](https://developers.cloudflare.com/radar/get-started/first-request/) to create an API token.
## License
@ -525,4 +526,6 @@ _PhishTank is either trademark or registered trademark of Cisco Systems, Inc._
[phishunt.io](https://phishunt.io/): All rights reserved by [Daniel López](https://twitter.com/0xDanielLopez)
[Cloudflare Radar](https://developers.cloudflare.com/radar/investigate/domain-ranking-datasets/): Available to free Cloudflare account
This repository is not endorsed by PhishTank/OpenDNS and OpenPhish.

View File

@ -1,11 +1,11 @@
#!/bin/sh
# works best on busybox sh
# works best on busybox ash
set -efx -o pipefail
alias curl="curl -L"
alias rm="rm -f"
alias rm="rm -rf"
## Use GNU grep, busybox grep is too slow
. "/etc/os-release"
@ -66,6 +66,33 @@ curl "https://tranco-list.eu/top-1m.csv.zip" -o "top-1m-tranco.zip"
bunzip2 -kc "phishtank.bz2" > "phishtank.csv"
## Cloudflare Radar
if [ -n "$CF_API" ]; then
mkdir -p "cf/"
# Get the latest domain ranking buckets
curl -X GET "https://api.cloudflare.com/client/v4/radar/datasets?limit=5&offset=0&datasetType=RANKING_BUCKET&format=json" \
-H "Authorization: Bearer $CF_API" -o "cf/datasets.json"
# Get the top 1m bucket's dataset ID
DATASET_ID=$(jq ".result.datasets[] | select(.meta.top==1000000) | .id" "cf/datasets.json")
# Get the dataset download url
curl --request POST \
--url "https://api.cloudflare.com/client/v4/radar/datasets/download" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer $CF_API" \
--data "{ \"datasetId\": $DATASET_ID }" \
-o "cf/dataset-url.json"
DATASET_URL=$(jq ".result.dataset.url" "cf/dataset-url.json" | sed 's/"//g')
curl -L "$DATASET_URL" -o "cf/top-1m-radar.zip"
## Parse the Radar 1 Million
unzip -p "cf/top-1m-radar.zip" | \
dos2unix | \
tr "[:upper:]" "[:lower:]" | \
grep -F "." | \
sed "s/^www\.//g" | \
sort -u > "top-1m-radar.txt"
fi
## Parse URLs
cat "phishtank.csv" | \
@ -146,10 +173,16 @@ sort -u > "top-1m-tranco.txt"
# ## Append new line https://unix.stackexchange.com/a/31955
# sed '$a\' > "oisd-exclude.txt"
# Merge Umbrella, Traco and self-maintained top domains
# Merge Umbrella, Tranco, Radar and self-maintained top domains
cat "top-1m-umbrella.txt" "top-1m-tranco.txt" "exclude.txt" | \
sort -u > "top-1m-well-known.txt"
if [ -n "$CF_API" ] && [ -f "top-1m-radar.txt" ]; then
cat "top-1m-radar.txt" >> "top-1m-well-known.txt"
# sort in-place
sort "top-1m-well-known.txt" -u -o "top-1m-well-known.txt"
fi
## Parse popular domains
cat "phishing-domains.txt" | \
@ -374,7 +407,7 @@ sed "2s/Domains Blocklist/Hosts Blocklist (IE)/" > "../public/phishing-filter.tp
## Clean up artifacts
rm "phishtank.csv" "top-1m-umbrella.zip" "top-1m-umbrella.txt" "top-1m-tranco.txt" "openphish-raw.txt" "phishunt.csv"
rm "phishtank.csv" "top-1m-umbrella.zip" "top-1m-umbrella.txt" "top-1m-tranco.txt" "openphish-raw.txt" "phishunt.csv" "cf/" "top-1m-radar.txt"
cd ../